78 EXTENT OF THE BEAJ-MANDAL, In the VaraM Purana, or rather in the interpolated section of that work known as the Mathura Mahatmya, the Mathura Mandal is described as twenty yojanas in extent. U n " My Mathura circle is one of twenty yojanas ; by bathing at any place therein a man is redeemed from all his sins." And taking the yojana as 7 miles and the kos as 1| mile, 20 yojanas would be nearly equal to 84 kos, the popular estimate of the distance travelled by the pilgrims in performing the Pari-krama, or £ perambulation' of Braj, It is pro- bable that if an accurate measurement were made, this would be found a very rough approximation to the actual length of the way ; though liberal allowance must be made for the constant ins and outs, turns and returns, which ultimately result in the circuit of a not very wide-spread area. There can be no doubt that the number 84, which in ancient Indian territorial divisions occurs as fre- quently as a hundred in English counties, and which enters largely into every cycle of Hindu legend and cosmogony, was originally selected for such general adoption as being the multiple of the number of months in the year with the number of days in the week. It is therefore peculiarly appropriate in connec- tion with the Braj Mandal ; if Krishna, in whose honour the perambulation is performed, be regarded as the Indian Apollo, or Sun-God. Thus, the magnifi- cent temple in Kashmir, dedicated to the sun tinder the title of Martand, has a colonnade of exactly 84 pillars,* It is sometimes said that the circle originally must have been of wider extent than now, since the city of Mathura, which is described as its centre, is more than 30 miles distant from the most northern point, Kotban, and only six from Ttim to tlie south ; and Elliot in his glossary quotes the following couplet as Ixing its limits : — si nw ira^r srii n " On one side Bar, on another Sona. on the tBird the town of Surasen ; these are the limits of the B-aj C%aurasi? the Mathura circle." •Mr. Fergnsson; in bis Indian ArcMtectnret doubts whether this temple was ever reallj dedi- cated to the sun, la so doing lie only betrays bis wonted linguistic ignorance. Martacd is not, as be supposes, simply a place-name, without any known connoiatioo, but is the actual dedi* catkwi title of the temple itself.